1% Tenofovir gel was shown to achieve an overall efficacy of 39% in the CAPRISA 004 study compared to women receiving placebo gel and 54% efficacy in women who had >80% adherence to gel use (65). Since only 54% efficacy in preventing HIV acquisition was achieved in highly adherent women, this suggests that other factors, apart from lack of adherence, are contributing to the poor efficacy of 1% tenofovir gel. It has become apparent that biological factors including the presence of genital inflammation (90) due to dysbiotic or suboptimal vaginal microbiota, as exemplified by bacterial vaginosis, as well as the presence of symptomatic and asymptomatic STIs contribute to attenuating the anti-HIV efficacy of 1% tenofovir gel.
In a post hoc prospective analysis of data from the CAPRISA 004 study, 1% tenofovir gel was 57% protective against HIV in women without genital inflammation, as determined by concentrations of nine pro-inflammatory cytokines in cervicovaginal lavages, and only 3% protective if genital inflammation was present (90). The role of dysbiotic or suboptimal vaginal microbiota associated with genital inflammation and driving increased HIV risk was demonstrated in a prospective study in adolescent girls and young women in sub-Saharan Africa (91). Individuals who were colonised with a highly diverse vaginal microbiota had just over a 4 fold increased risk of acquiring HIV infection from their male partners compared to women colonised with a beneficial Lactobacillus species (L. crispatus) (91). These women also had significantly increased levels of pro-inflammatory cytokines and chemokines in genital secretions and an increase in activated CD4+ T cells in the cervix providing a plausible mechanism for increased HIV risk (91) that could attenuate the efficacy of topical PrEP.
In addition, compared to women with a Lactobacillus-dominated vaginal microbiota, women with bacterial vaginosis have disrupted cervicovaginal epithelial barrier integrity that can promote penetration of HIV through the epithelial barrier to infect subepithelial HIV target cells. In addition, women with bacterial vaginosis have adverse changes in the protective properties of cervical mucous which acts as a physical barrier to prevent HIV entering the cervix (92, 93).
A dysbiotic vaginal microbiota has also been shown to modulate the efficacy of 1% tenofovir gel in the CAPRISA 004 study by metabolising tenofovir (94). This study showed that tenofovir-reduced HIV incidence by 61% in women with Lactobacillus-dominated microbiota and only by 18% in women colonised with a highly diverse vaginal microbiota (i.e. non-Lactobacillus bacteria dominated) (94). Mucosal tenofovir levels were lower in women with non-Lactobacillus bacteria and these bacteria were responsible for metabolising tenofovir (94). Studies are ongoing to determine if other ARVs used for topical PrEP (e.g. dapivirine) are also metabolised in a dysbiotic microbiota. In contrast, the efficacy of oral daily PrEP (either tenofovir monotherapy or Truvada) was not significantly different among African women with healthy vs abnormal vaginal microbiota as defined by Nugent Score (95). However, not all adolescent girls and young women in South Africa with a diverse vaginal microbiota are positive by the Nugent method (91). Further studies will be needed to determine whether the vaginal microbiota can impact on the efficacy of oral PrEP.
Collectively these studies have prompted an interest in strategies to reduce genital inflammation, including modifying a dysbiotic vaginal microbiota by the use of antibiotics, pre/probiotics or postbiotics to augment HIV prevention efforts (22) (96) (40).